Nekomata 14B Instruction
This model instruction-tunes Nekomata 14B, retaining its Qwen ancestry and using the Alpaca task format.
Instruction data adds task-following behavior
The card lists Dolly, Japanese Dolly, FLAN and selected Izumi Lab bilingual datasets. The original Nekomata tokenizer is retained. Official GGUF files are packaging variants of this release.Nekomata 14B Instruction official model card
Check gains on the intended Japanese task
Practical guidance. Compare the 14B and 7B instruction releases on the same Japanese task. A larger decoder can need more memory without improving a narrow extraction rule, so assess the result you actually need.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | nekomata-14b-instructionnekomata-14b-instruction-gguf |
| License | Tongyi Qianwen License Agreement.Nekomata 14B Instruction official model card |
| Architecture | 40 layers with hidden width 5,120.Nekomata 14B Instruction official model card |
Related models
- Nekomata 14B
rinna · Nekomata
Nekomata 14B continues pretraining Qwen-14B on about 66B tokens of Japanese and English text.
Launch date not confirmedResearch releaseDownloadable - Nekomata 7B Instruction
rinna · Nekomata
This model instruction-tunes Nekomata 7B, retaining its Qwen ancestry and using the Alpaca task format.
Launch date not confirmedDownloadable